A novel retrovenous myocardial drug delivery system has been developed and Phase I feasibility demonstrated in sheep with acute LAD coronary artery occlusion. Control sheep with intravenous lidocaine therapy developed ventricular fibrillation in a mean time of 12.5 minutes after PTCA-LAD occlusion. In contrast, retrovenously treated sheep with an identical lidocaine dose and venous blood survived prolonged study periods to 3 hours despite infarction in the area-at-risk, severe ECG changes, and myocardial depression. Phase II work will determine the efficacy of combined retrovenous arterialized blood and lidocaine for prevention of arrhythmias and salvage of ischemic myocardium. A dose-response of retrovenous lidocaine after coronary artery occlusion will be obtained. Four-lumen retrovenous catheters will be fabricated and tested to evaluate our hypothesis that recirculation of therapeutic agents via the coronary sinus reduces the effective dose and thus the proarrhythmic or toxic effects of the therapeutic agents. A second hypothesis will be tested in a model of acute coronary thrombosis that retrovenous t-PA thrombolytic therapy by itself or in combination with selective arterial or intravenous therapy accelerates clot lysis and time of myocardial reperfusion. The retrovenous drive system developed during Phase I will be modified for controlled drug delivery with safeguards and alarms for Phase III clinical application.